Average Order Value: Formula, Google Ads Mismatch and ROAS

Average order value (AOV) is revenue divided by the number of orders. For ad traffic, Google Ads shows it in the Value / conv. column, which divides conversion value by conversions. That figure rarely matches your shop’s. Google Ads counts only the orders it credits to its ads, can date them differently and uses whatever amount your tag sends. While conversion rate and cost per click stay the same, AOV moves ROAS by the same percentage: ROAS = conversion rate × AOV ÷ cost per click.

What is average order value?

Average order value is how much one order brings in on average. Shopify’s guide (Average Order Value: Formula and 7 Ways) gives the basic formula:

AOV = revenue ÷ number of orders

AOV is one of three numbers behind ROAS. The other two are cost per click and conversion rate, the share of clicks that end in an order. The guide to ecommerce unit economics shows how the three combine with your margin into a break-even point.

Example store, not client data.

A tableware shop with 3,000 products gets 300 orders from Google Ads in a month, worth 180,000 in total. Its average order value is 180,000 ÷ 300 = 600.

AOV measures money per order. Google Ads reports items per order separately, as average cart size (units sold ÷ orders), once you send cart data with your purchases.

How do you calculate average order value for ad traffic?

Google Ads already does the division. According to the Google Ads API reference (metrics), Value / conv. is your total conversion value divided by the number in the Conversions column. The Conversions column counts only your primary conversion actions (Conversion goals). So Value / conv. is your average order value only when your primary action is a purchase.

1. Add the columns. In the campaigns table, show Conversions, Conv. value and Value / conv. for the account and for each campaign.
2. Use Conversions, not All conv. Secondary actions stay out of Conversions but are still counted in All conv. If you read All conv., an add-to-cart that carries a value gets mixed into your “order” value.
3. Choose how orders are dated. Google Ads has a second version of the column, Value / conv. (by conv. time). When you split the report by day, it puts each order on the day the order happened. That is closer to how your shop counts.
4. Check the cart-data version if you have it. With cart data, Google Ads also reports average order value as revenue ÷ orders. The same API reference defines revenue here as item prices minus discounts. Because it comes from the item prices in the cart data, it can differ from the amount in your tag’s value field.
5. Take the matching figure from your shop. Use only orders that came from Google Ads, for the same dates. In Google Analytics, filter reports to session source / medium google / cpc.

Why is your Google Ads AOV different from your shop’s?

Your shop divides all its revenue by all its orders. Google Ads divides the value it credits to ads by the conversions it credits to ads. So expect some gap even when tracking works: the two sides count different orders and date them differently.

These are the usual causes:

The Google Ads API reference lists the attribution models in AttributionModel and the two counting options in ConversionActionCountingType.

A default value flattens AOV completely, because every order arrives with the same amount. According to the API reference (ValueSettings), Google Ads uses the default value when a conversion arrives with a missing or invalid value. The same happens when the action is set to always use the default. If your products have different prices, each purchase has to carry its own order amount and currency.

Double counting is harder to spot through AOV. If both duplicate actions carry the real amount, orders and value double together and AOV looks normal, while ROAS doubles. So compare the number of conversions with your orders as well as the average. Send a transaction ID with every purchase, and keep only purchase actions as primary.

Returns need a separate upload. According to the Google Ads API guide (Import conversion adjustments), you can adjust a conversion after it has been reported. You can retract a returned or cancelled order, or restate a partly returned one at a lower value. Without that upload, Google Ads keeps the amount reported at the time of purchase.

Should average order value include VAT, delivery and discounts?

There is no single correct definition. Shopify’s formula says “revenue” and leaves the contents to you. Google’s cart-data AOV adds up item prices and takes off discounts. What matters is that both figures you compare use the same definition.

The value inside Google Ads matters more than the label in your report. Your ROAS and your bid targets rest on whatever the tag sends. Whether that amount should carry VAT and delivery is a separate question: the choice shifts the ROAS your bids aim for. We cover it in what conversion value should include.

Why can one large order shift the average?

AOV is a mean, and a few large values can pull a mean up. Shopify’s guide quotes a practitioner, Taylor Holiday, who advises looking at the mean, the median and the most common order value together. The guide’s example shows a handful of high-value purchases pulling the average above what most customers spend.

Example store, not client data.

Take the tableware shop’s month of 300 orders worth 180,000 and add one wholesale order of 60,000. Revenue becomes 240,000 from 301 orders, and AOV rises from 600 to about 797, a third higher. The median barely moves: one added order shifts it by at most one place in the sorted list of 301 orders.

The portal’s order-value checks set such outliers aside before using the account’s average order value as a yardstick.

How much does average order value move ROAS?

ROAS = conversion rate × AOV ÷ cost per click. While the other two numbers stay the same, AOV and ROAS move by the same percentage.

Example store, not client data.

The tableware shop converts 3.75% of clicks, pays 5 per click and has an AOV of 600. Its ROAS is 3.75% × 600 ÷ 5 = 4.5, or 450% as a percentage. Raise AOV by 10% to 660 and keep the other two the same: ROAS becomes 4.95, also 10% higher.

In practice, the other two numbers rarely stay the same. A higher price can lower your conversion rate. A higher AOV also raises the most you can pay for a click without losing money: your break-even cost per click.

In our data, ROAS follows AOV more closely within a store than between stores

Between stores, ROAS stopped rising after the middle segment

Our study of 1.4 million products grouped 128 stores by average order value, using 13 months of data. Loss share is the share of budget spent on products that made no sales in that time.

Three things stand out:

• ROAS rose from the low to the middle segment, then stopped. It went from 300% to 444%, and the high segment sat at 418%. The premium figure of 1118% comes from only 7 stores, and the study warns against judging the premium segment from them.
• Higher AOV came with more budget on products that never sold. Loss share rose from 32.6% in the low segment to 39.1% in the middle and 53.4% in the high one. The premium segment sat at 48.0%. The share of products with sales fell from 32.2% to 13.7%.
• Clicks to purchase did not grow with price. A purchase took 34 clicks in the low segment and 22 in the high one.

Within one store, the link is clearer than between stores

We looked at 1,360 store-months of GetProfit data from June 2025 to June 2026. We measured each store’s monthly change against the median store in the same month, which strips out season and holidays. Changes in ROAS went together with changes in AOV (correlation +0.540) about as closely as with conversion rate (+0.527). Their link with cost per click was weak (−0.100).

Between stores, the link was weaker. Across 114 stores with at least eight months of data in the same window, AOV and ROAS correlated at +0.329. Revenue per click, which combines AOV and conversion rate in one number, tracked ROAS at +0.633.

These are observations, not an experiment: they show what moves together, not what causes what. In practice, compare your AOV with your own earlier months rather than with other stores. To choose which lever to work on first, AOV or conversion rate, see how to increase ROAS.

When Value / conv. jumps or drops, check the value first

Sometimes Value / conv. changes sharply while your shop’s AOV for the same days stays flat. Then the change lies in what Google Ads receives rather than in what customers buy.

The portal’s conversion check runs this comparison per campaign. Once a Search, Shopping or Performance Max campaign has 10 conversions, the portal compares its AOV with the account’s average order value. It flags a campaign whose average is far above what a real order could be worth. It also catches the case where your settings say “pass the real amount”, yet every order still arrives with one and the same number.

A second check looks at single amounts. It catches a product whose conversion value for one day is more than 30 times the account’s average order value. The portal puts each such case on a separate list with the date, campaign and product.

What to do with your average order value

1. Take the last full month in Google Ads. Add Conversions, Conv. value and Value / conv., and check that your primary conversion action is a purchase.
2. Take the same month from your shop. Use only orders from Google Ads, and write down what your figure includes: VAT, delivery, discounts, returns.
3. Line up the dates. Compare your shop’s figure with Value / conv. (by conv. time), because your shop counts orders on the day they happened.
4. If the gap is wide, check the usual causes in order. Start with the default value and duplicate purchase actions, then VAT, delivery and returns.
5. Look at the median as well as the mean. List the month’s largest orders. In the tableware shop example, one wholesale order lifted AOV by a third.
6. Track AOV by campaign against your own past months. In our data, ROAS moved with AOV more closely within a store than between stores.
7. Before you push AOV up with prices or bundles, recheck conversion rate and break-even cost per click. A higher AOV with a lower conversion rate can leave ROAS where it was.

See what share of your numbers you can trust. Sign in with Google in one click. The portal changes nothing without your consent.

Check your conversions →

Sources

• metrics — Google Ads API — Value / conv. = value of conversions ÷ number of conversions; the (by conv. time) columns use the conversion date; with cart data, average order value = revenue ÷ orders, revenue = item prices minus discounts, average cart size = products sold ÷ orders. Checked 2 October 2026.
• Conversion goals — Google Ads API — primary actions are reported in Conversions; secondary actions only in All conv. Checked 2 October 2026.
• Conversion reporting — Google Ads API — the names of the Conversions, Conv. value, Value / conv. and (by conv. time) columns in the Google Ads interface. Checked 2 October 2026.
• AttributionModel — Google Ads API — data-driven attribution distributes credit for a conversion among clicks. Checked 2 October 2026.
• ConversionActionCountingType — Google Ads API — one conversion per click versus all conversions per click. Checked 2 October 2026.
• ValueSettings — Google Ads API — the default value is used for a missing or invalid value, or when the action always uses it. Checked 2 October 2026.
• Import conversion adjustments — Google Ads API — a reported conversion can later be retracted or restated. Checked 2 October 2026.
• Average Order Value: Formula and 7 Ways (2026) — AOV = total revenue ÷ number of orders; a practitioner’s advice to read mean, median and mode together. Checked 2 October 2026.
• GetProfit study: 1,404,808 products, 130+ stores, 13 months; 128 stores split by average order value — ROAS, loss share, products with sales and clicks to purchase by segment.
• GetProfit data: 1,360 store-months, June 2025 – June 2026, each measured against the median store in the same month — how ROAS moved with AOV, conversion rate and cost per click; 114 stores with at least eight months — AOV, revenue per click and ROAS between stores.
• GetProfit portal methodology — the campaign order-value check from 10 conversions in Search, Shopping and Performance Max campaigns; the 30× limit for one product’s conversion value in a day; outliers set aside before the account’s average order is used as a yardstick.